Despite dramatic improvement in cure rates of childhood acute lymphoblastic leukemia (ALL), stark racial disparities have persisted in both ALL susceptibility and treatment outcome, e.g. Hispanic children have the highest incidence of ALL and the poorest survival among major race/ethnic groups in the US. While underlying causes of such racial disparities are largely unknown, there is a particular paucity of basic science studies of biological differences in ALL by race. Without substantive investment in translational research of ALL disparity, this catastrophic disease will continue to disproportionally affect Hispanic children and their families. Objective/hypothesis Recent genomic studies of ALL by our group has established a genetic basis for racial disparities in ALL outcome, e.g. Native American genetic ancestry is strongly correlated with relapse rate (Nat Genet 43:237). In particular, ancestry-related genetic variation in ARID5B contributes significantly to the increased ALL incidence and relapse risk in Hispanic children (J Clin Oncol 30:751). Our further mechanistic studies linked ARID5B to response of ALL cells to methotrexate and 6-mercaptopurine, backbone of almost all contemporary ALL treatment regimens. Building upon these preliminary data, we hypothesize that ARID5B is a critical determinant of racial differences in ALL treatment outcome, via its effects on the disposition of and response to antileukemic drugs. The objectives of this project are to comprehensively identify pharmacogenetic variants in ARID5B and to mechanistically describe how ARID5B affects ALL drug response. Approach Taking a comprehensive approach, this project combines patient-oriented pharmacogenetic studies and laboratory-based molecular pharmacology experiments. The goal is to not only to discover clinically relevant prognostic markers in ALL but also to understand the biology from which the prognostic associations arise. Thus, we will first resequence ARID5B in a multiethnic group of children with ALL, followed by pharmacogenetic association studies of ARID5B in state-of-the-art Children's Oncology Group ALL trials (>5,000 children with ALL). Finally, we will mechanistically characterize ARID5B's effects on methotrexate and 6-mercaptopurine disposition and response, using ALL cell lines, mouse models, and in children with ALL. Impact and significance Successful completion of these studies is likely to establish novel biological mechanisms responsible for racial disparities in AL and enable the development of therapeutic approaches to overcome racial gaps. The long-term goal of our research is to characterize genomic features of ALL across race groups, and to implement pharmacogenomics-guided treatment individualization to improve outcome of ALL for all children.